2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-30 08:04:13 +08:00
linux-next/drivers/base/node.c

713 lines
19 KiB
C
Raw Normal View History

/*
* Basic Node interface support
*/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/mm.h>
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
#include <linux/memory.h>
#include <linux/vmstat.h>
#include <linux/notifier.h>
#include <linux/node.h>
#include <linux/hugetlb.h>
#include <linux/compaction.h>
#include <linux/cpumask.h>
#include <linux/topology.h>
#include <linux/nodemask.h>
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
#include <linux/cpu.h>
#include <linux/device.h>
#include <linux/swap.h>
#include <linux/slab.h>
static struct bus_type node_subsys = {
.name = "node",
.dev_name = "node",
};
static ssize_t node_read_cpumap(struct device *dev, bool list, char *buf)
{
struct node *node_dev = to_node(dev);
const struct cpumask *mask = cpumask_of_node(node_dev->dev.id);
/* 2008/04/07: buf currently PAGE_SIZE, need 9 chars per 32 bits. */
BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1));
return cpumap_print_to_pagebuf(list, buf, mask);
}
static inline ssize_t node_read_cpumask(struct device *dev,
struct device_attribute *attr, char *buf)
{
return node_read_cpumap(dev, false, buf);
}
static inline ssize_t node_read_cpulist(struct device *dev,
struct device_attribute *attr, char *buf)
{
return node_read_cpumap(dev, true, buf);
}
static DEVICE_ATTR(cpumap, S_IRUGO, node_read_cpumask, NULL);
static DEVICE_ATTR(cpulist, S_IRUGO, node_read_cpulist, NULL);
#define K(x) ((x) << (PAGE_SHIFT - 10))
static ssize_t node_read_meminfo(struct device *dev,
struct device_attribute *attr, char *buf)
{
int n;
int nid = dev->id;
mm, vmscan: move LRU lists to node This moves the LRU lists from the zone to the node and related data such as counters, tracing, congestion tracking and writeback tracking. Unfortunately, due to reclaim and compaction retry logic, it is necessary to account for the number of LRU pages on both zone and node logic. Most reclaim logic is based on the node counters but the retry logic uses the zone counters which do not distinguish inactive and active sizes. It would be possible to leave the LRU counters on a per-zone basis but it's a heavier calculation across multiple cache lines that is much more frequent than the retry checks. Other than the LRU counters, this is mostly a mechanical patch but note that it introduces a number of anomalies. For example, the scans are per-zone but using per-node counters. We also mark a node as congested when a zone is congested. This causes weird problems that are fixed later but is easier to review. In the event that there is excessive overhead on 32-bit systems due to the nodes being on LRU then there are two potential solutions 1. Long-term isolation of highmem pages when reclaim is lowmem When pages are skipped, they are immediately added back onto the LRU list. If lowmem reclaim persisted for long periods of time, the same highmem pages get continually scanned. The idea would be that lowmem keeps those pages on a separate list until a reclaim for highmem pages arrives that splices the highmem pages back onto the LRU. It potentially could be implemented similar to the UNEVICTABLE list. That would reduce the skip rate with the potential corner case is that highmem pages have to be scanned and reclaimed to free lowmem slab pages. 2. Linear scan lowmem pages if the initial LRU shrink fails This will break LRU ordering but may be preferable and faster during memory pressure than skipping LRU pages. Link: http://lkml.kernel.org/r/1467970510-21195-4-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Rik van Riel <riel@surriel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:31 +08:00
struct pglist_data *pgdat = NODE_DATA(nid);
struct sysinfo i;
si_meminfo_node(&i, nid);
n = sprintf(buf,
vmscan: split LRU lists into anon & file sets Split the LRU lists in two, one set for pages that are backed by real file systems ("file") and one for pages that are backed by memory and swap ("anon"). The latter includes tmpfs. The advantage of doing this is that the VM will not have to scan over lots of anonymous pages (which we generally do not want to swap out), just to find the page cache pages that it should evict. This patch has the infrastructure and a basic policy to balance how much we scan the anon lists and how much we scan the file lists. The big policy changes are in separate patches. [lee.schermerhorn@hp.com: collect lru meminfo statistics from correct offset] [kosaki.motohiro@jp.fujitsu.com: prevent incorrect oom under split_lru] [kosaki.motohiro@jp.fujitsu.com: fix pagevec_move_tail() doesn't treat unevictable page] [hugh@veritas.com: memcg swapbacked pages active] [hugh@veritas.com: splitlru: BDI_CAP_SWAP_BACKED] [akpm@linux-foundation.org: fix /proc/vmstat units] [nishimura@mxp.nes.nec.co.jp: memcg: fix handling of shmem migration] [kosaki.motohiro@jp.fujitsu.com: adjust Quicklists field of /proc/meminfo] [kosaki.motohiro@jp.fujitsu.com: fix style issue of get_scan_ratio()] Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 11:26:32 +08:00
"Node %d MemTotal: %8lu kB\n"
"Node %d MemFree: %8lu kB\n"
"Node %d MemUsed: %8lu kB\n"
"Node %d Active: %8lu kB\n"
"Node %d Inactive: %8lu kB\n"
"Node %d Active(anon): %8lu kB\n"
"Node %d Inactive(anon): %8lu kB\n"
"Node %d Active(file): %8lu kB\n"
"Node %d Inactive(file): %8lu kB\n"
"Node %d Unevictable: %8lu kB\n"
"Node %d Mlocked: %8lu kB\n",
nid, K(i.totalram),
nid, K(i.freeram),
nid, K(i.totalram - i.freeram),
mm, vmscan: move LRU lists to node This moves the LRU lists from the zone to the node and related data such as counters, tracing, congestion tracking and writeback tracking. Unfortunately, due to reclaim and compaction retry logic, it is necessary to account for the number of LRU pages on both zone and node logic. Most reclaim logic is based on the node counters but the retry logic uses the zone counters which do not distinguish inactive and active sizes. It would be possible to leave the LRU counters on a per-zone basis but it's a heavier calculation across multiple cache lines that is much more frequent than the retry checks. Other than the LRU counters, this is mostly a mechanical patch but note that it introduces a number of anomalies. For example, the scans are per-zone but using per-node counters. We also mark a node as congested when a zone is congested. This causes weird problems that are fixed later but is easier to review. In the event that there is excessive overhead on 32-bit systems due to the nodes being on LRU then there are two potential solutions 1. Long-term isolation of highmem pages when reclaim is lowmem When pages are skipped, they are immediately added back onto the LRU list. If lowmem reclaim persisted for long periods of time, the same highmem pages get continually scanned. The idea would be that lowmem keeps those pages on a separate list until a reclaim for highmem pages arrives that splices the highmem pages back onto the LRU. It potentially could be implemented similar to the UNEVICTABLE list. That would reduce the skip rate with the potential corner case is that highmem pages have to be scanned and reclaimed to free lowmem slab pages. 2. Linear scan lowmem pages if the initial LRU shrink fails This will break LRU ordering but may be preferable and faster during memory pressure than skipping LRU pages. Link: http://lkml.kernel.org/r/1467970510-21195-4-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Rik van Riel <riel@surriel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:31 +08:00
nid, K(node_page_state(pgdat, NR_ACTIVE_ANON) +
node_page_state(pgdat, NR_ACTIVE_FILE)),
nid, K(node_page_state(pgdat, NR_INACTIVE_ANON) +
node_page_state(pgdat, NR_INACTIVE_FILE)),
nid, K(node_page_state(pgdat, NR_ACTIVE_ANON)),
nid, K(node_page_state(pgdat, NR_INACTIVE_ANON)),
nid, K(node_page_state(pgdat, NR_ACTIVE_FILE)),
nid, K(node_page_state(pgdat, NR_INACTIVE_FILE)),
nid, K(node_page_state(pgdat, NR_UNEVICTABLE)),
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
nid, K(sum_zone_node_page_state(nid, NR_MLOCK)));
#ifdef CONFIG_HIGHMEM
n += sprintf(buf + n,
vmscan: split LRU lists into anon & file sets Split the LRU lists in two, one set for pages that are backed by real file systems ("file") and one for pages that are backed by memory and swap ("anon"). The latter includes tmpfs. The advantage of doing this is that the VM will not have to scan over lots of anonymous pages (which we generally do not want to swap out), just to find the page cache pages that it should evict. This patch has the infrastructure and a basic policy to balance how much we scan the anon lists and how much we scan the file lists. The big policy changes are in separate patches. [lee.schermerhorn@hp.com: collect lru meminfo statistics from correct offset] [kosaki.motohiro@jp.fujitsu.com: prevent incorrect oom under split_lru] [kosaki.motohiro@jp.fujitsu.com: fix pagevec_move_tail() doesn't treat unevictable page] [hugh@veritas.com: memcg swapbacked pages active] [hugh@veritas.com: splitlru: BDI_CAP_SWAP_BACKED] [akpm@linux-foundation.org: fix /proc/vmstat units] [nishimura@mxp.nes.nec.co.jp: memcg: fix handling of shmem migration] [kosaki.motohiro@jp.fujitsu.com: adjust Quicklists field of /proc/meminfo] [kosaki.motohiro@jp.fujitsu.com: fix style issue of get_scan_ratio()] Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 11:26:32 +08:00
"Node %d HighTotal: %8lu kB\n"
"Node %d HighFree: %8lu kB\n"
"Node %d LowTotal: %8lu kB\n"
"Node %d LowFree: %8lu kB\n",
nid, K(i.totalhigh),
nid, K(i.freehigh),
nid, K(i.totalram - i.totalhigh),
nid, K(i.freeram - i.freehigh));
#endif
n += sprintf(buf + n,
vmscan: split LRU lists into anon & file sets Split the LRU lists in two, one set for pages that are backed by real file systems ("file") and one for pages that are backed by memory and swap ("anon"). The latter includes tmpfs. The advantage of doing this is that the VM will not have to scan over lots of anonymous pages (which we generally do not want to swap out), just to find the page cache pages that it should evict. This patch has the infrastructure and a basic policy to balance how much we scan the anon lists and how much we scan the file lists. The big policy changes are in separate patches. [lee.schermerhorn@hp.com: collect lru meminfo statistics from correct offset] [kosaki.motohiro@jp.fujitsu.com: prevent incorrect oom under split_lru] [kosaki.motohiro@jp.fujitsu.com: fix pagevec_move_tail() doesn't treat unevictable page] [hugh@veritas.com: memcg swapbacked pages active] [hugh@veritas.com: splitlru: BDI_CAP_SWAP_BACKED] [akpm@linux-foundation.org: fix /proc/vmstat units] [nishimura@mxp.nes.nec.co.jp: memcg: fix handling of shmem migration] [kosaki.motohiro@jp.fujitsu.com: adjust Quicklists field of /proc/meminfo] [kosaki.motohiro@jp.fujitsu.com: fix style issue of get_scan_ratio()] Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 11:26:32 +08:00
"Node %d Dirty: %8lu kB\n"
"Node %d Writeback: %8lu kB\n"
"Node %d FilePages: %8lu kB\n"
"Node %d Mapped: %8lu kB\n"
"Node %d AnonPages: %8lu kB\n"
"Node %d Shmem: %8lu kB\n"
"Node %d KernelStack: %8lu kB\n"
vmscan: split LRU lists into anon & file sets Split the LRU lists in two, one set for pages that are backed by real file systems ("file") and one for pages that are backed by memory and swap ("anon"). The latter includes tmpfs. The advantage of doing this is that the VM will not have to scan over lots of anonymous pages (which we generally do not want to swap out), just to find the page cache pages that it should evict. This patch has the infrastructure and a basic policy to balance how much we scan the anon lists and how much we scan the file lists. The big policy changes are in separate patches. [lee.schermerhorn@hp.com: collect lru meminfo statistics from correct offset] [kosaki.motohiro@jp.fujitsu.com: prevent incorrect oom under split_lru] [kosaki.motohiro@jp.fujitsu.com: fix pagevec_move_tail() doesn't treat unevictable page] [hugh@veritas.com: memcg swapbacked pages active] [hugh@veritas.com: splitlru: BDI_CAP_SWAP_BACKED] [akpm@linux-foundation.org: fix /proc/vmstat units] [nishimura@mxp.nes.nec.co.jp: memcg: fix handling of shmem migration] [kosaki.motohiro@jp.fujitsu.com: adjust Quicklists field of /proc/meminfo] [kosaki.motohiro@jp.fujitsu.com: fix style issue of get_scan_ratio()] Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-19 11:26:32 +08:00
"Node %d PageTables: %8lu kB\n"
"Node %d NFS_Unstable: %8lu kB\n"
"Node %d Bounce: %8lu kB\n"
"Node %d WritebackTmp: %8lu kB\n"
"Node %d Slab: %8lu kB\n"
"Node %d SReclaimable: %8lu kB\n"
"Node %d SUnreclaim: %8lu kB\n"
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
"Node %d AnonHugePages: %8lu kB\n"
"Node %d ShmemHugePages: %8lu kB\n"
"Node %d ShmemPmdMapped: %8lu kB\n"
#endif
,
nid, K(node_page_state(pgdat, NR_FILE_DIRTY)),
nid, K(node_page_state(pgdat, NR_WRITEBACK)),
nid, K(node_page_state(pgdat, NR_FILE_PAGES)),
nid, K(node_page_state(pgdat, NR_FILE_MAPPED)),
nid, K(node_page_state(pgdat, NR_ANON_MAPPED)),
nid, K(i.sharedram),
nid, sum_zone_node_page_state(nid, NR_KERNEL_STACK_KB),
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
nid, K(sum_zone_node_page_state(nid, NR_PAGETABLE)),
nid, K(node_page_state(pgdat, NR_UNSTABLE_NFS)),
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
nid, K(sum_zone_node_page_state(nid, NR_BOUNCE)),
nid, K(node_page_state(pgdat, NR_WRITEBACK_TEMP)),
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
nid, K(sum_zone_node_page_state(nid, NR_SLAB_RECLAIMABLE) +
sum_zone_node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
nid, K(sum_zone_node_page_state(nid, NR_SLAB_RECLAIMABLE)),
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
nid, K(sum_zone_node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
nid, K(node_page_state(pgdat, NR_ANON_THPS) *
HPAGE_PMD_NR),
nid, K(node_page_state(pgdat, NR_SHMEM_THPS) *
HPAGE_PMD_NR),
nid, K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED) *
HPAGE_PMD_NR));
#else
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
nid, K(sum_zone_node_page_state(nid, NR_SLAB_UNRECLAIMABLE)));
#endif
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
}
#undef K
static DEVICE_ATTR(meminfo, S_IRUGO, node_read_meminfo, NULL);
static ssize_t node_read_numastat(struct device *dev,
struct device_attribute *attr, char *buf)
{
return sprintf(buf,
"numa_hit %lu\n"
"numa_miss %lu\n"
"numa_foreign %lu\n"
"interleave_hit %lu\n"
"local_node %lu\n"
"other_node %lu\n",
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
sum_zone_node_page_state(dev->id, NUMA_HIT),
sum_zone_node_page_state(dev->id, NUMA_MISS),
sum_zone_node_page_state(dev->id, NUMA_FOREIGN),
sum_zone_node_page_state(dev->id, NUMA_INTERLEAVE_HIT),
sum_zone_node_page_state(dev->id, NUMA_LOCAL),
sum_zone_node_page_state(dev->id, NUMA_OTHER));
}
static DEVICE_ATTR(numastat, S_IRUGO, node_read_numastat, NULL);
static ssize_t node_read_vmstat(struct device *dev,
struct device_attribute *attr, char *buf)
{
int nid = dev->id;
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
struct pglist_data *pgdat = NODE_DATA(nid);
int i;
int n = 0;
for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
n += sprintf(buf+n, "%s %lu\n", vmstat_text[i],
mm, vmstat: add infrastructure for per-node vmstats Patchset: "Move LRU page reclaim from zones to nodes v9" This series moves LRUs from the zones to the node. While this is a current rebase, the test results were based on mmotm as of June 23rd. Conceptually, this series is simple but there are a lot of details. Some of the broad motivations for this are; 1. The residency of a page partially depends on what zone the page was allocated from. This is partially combatted by the fair zone allocation policy but that is a partial solution that introduces overhead in the page allocator paths. 2. Currently, reclaim on node 0 behaves slightly different to node 1. For example, direct reclaim scans in zonelist order and reclaims even if the zone is over the high watermark regardless of the age of pages in that LRU. Kswapd on the other hand starts reclaim on the highest unbalanced zone. A difference in distribution of file/anon pages due to when they were allocated results can result in a difference in again. While the fair zone allocation policy mitigates some of the problems here, the page reclaim results on a multi-zone node will always be different to a single-zone node. it was scheduled on as a result. 3. kswapd and the page allocator scan zones in the opposite order to avoid interfering with each other but it's sensitive to timing. This mitigates the page allocator using pages that were allocated very recently in the ideal case but it's sensitive to timing. When kswapd is allocating from lower zones then it's great but during the rebalancing of the highest zone, the page allocator and kswapd interfere with each other. It's worse if the highest zone is small and difficult to balance. 4. slab shrinkers are node-based which makes it harder to identify the exact relationship between slab reclaim and LRU reclaim. The reason we have zone-based reclaim is that we used to have large highmem zones in common configurations and it was necessary to quickly find ZONE_NORMAL pages for reclaim. Today, this is much less of a concern as machines with lots of memory will (or should) use 64-bit kernels. Combinations of 32-bit hardware and 64-bit hardware are rare. Machines that do use highmem should have relatively low highmem:lowmem ratios than we worried about in the past. Conceptually, moving to node LRUs should be easier to understand. The page allocator plays fewer tricks to game reclaim and reclaim behaves similarly on all nodes. The series has been tested on a 16 core UMA machine and a 2-socket 48 core NUMA machine. The UMA results are presented in most cases as the NUMA machine behaved similarly. pagealloc --------- This is a microbenchmark that shows the benefit of removing the fair zone allocation policy. It was tested uip to order-4 but only orders 0 and 1 are shown as the other orders were comparable. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min total-odr0-1 490.00 ( 0.00%) 457.00 ( 6.73%) Min total-odr0-2 347.00 ( 0.00%) 329.00 ( 5.19%) Min total-odr0-4 288.00 ( 0.00%) 273.00 ( 5.21%) Min total-odr0-8 251.00 ( 0.00%) 239.00 ( 4.78%) Min total-odr0-16 234.00 ( 0.00%) 222.00 ( 5.13%) Min total-odr0-32 223.00 ( 0.00%) 211.00 ( 5.38%) Min total-odr0-64 217.00 ( 0.00%) 208.00 ( 4.15%) Min total-odr0-128 214.00 ( 0.00%) 204.00 ( 4.67%) Min total-odr0-256 250.00 ( 0.00%) 230.00 ( 8.00%) Min total-odr0-512 271.00 ( 0.00%) 269.00 ( 0.74%) Min total-odr0-1024 291.00 ( 0.00%) 282.00 ( 3.09%) Min total-odr0-2048 303.00 ( 0.00%) 296.00 ( 2.31%) Min total-odr0-4096 311.00 ( 0.00%) 309.00 ( 0.64%) Min total-odr0-8192 316.00 ( 0.00%) 314.00 ( 0.63%) Min total-odr0-16384 317.00 ( 0.00%) 315.00 ( 0.63%) Min total-odr1-1 742.00 ( 0.00%) 712.00 ( 4.04%) Min total-odr1-2 562.00 ( 0.00%) 530.00 ( 5.69%) Min total-odr1-4 457.00 ( 0.00%) 433.00 ( 5.25%) Min total-odr1-8 411.00 ( 0.00%) 381.00 ( 7.30%) Min total-odr1-16 381.00 ( 0.00%) 356.00 ( 6.56%) Min total-odr1-32 372.00 ( 0.00%) 346.00 ( 6.99%) Min total-odr1-64 372.00 ( 0.00%) 343.00 ( 7.80%) Min total-odr1-128 375.00 ( 0.00%) 351.00 ( 6.40%) Min total-odr1-256 379.00 ( 0.00%) 351.00 ( 7.39%) Min total-odr1-512 385.00 ( 0.00%) 355.00 ( 7.79%) Min total-odr1-1024 386.00 ( 0.00%) 358.00 ( 7.25%) Min total-odr1-2048 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-4096 390.00 ( 0.00%) 362.00 ( 7.18%) Min total-odr1-8192 388.00 ( 0.00%) 363.00 ( 6.44%) This shows a steady improvement throughout. The primary benefit is from reduced system CPU usage which is obvious from the overall times; 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 User 189.19 191.80 System 2604.45 2533.56 Elapsed 2855.30 2786.39 The vmstats also showed that the fair zone allocation policy was definitely removed as can be seen here; 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v8 DMA32 allocs 28794729769 0 Normal allocs 48432501431 77227309877 Movable allocs 0 0 tiobench on ext4 ---------------- tiobench is a benchmark that artifically benefits if old pages remain resident while new pages get reclaimed. The fair zone allocation policy mitigates this problem so pages age fairly. While the benchmark has problems, it is important that tiobench performance remains constant as it implies that page aging problems that the fair zone allocation policy fixes are not re-introduced. 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 Min PotentialReadSpeed 89.65 ( 0.00%) 90.21 ( 0.62%) Min SeqRead-MB/sec-1 82.68 ( 0.00%) 82.01 ( -0.81%) Min SeqRead-MB/sec-2 72.76 ( 0.00%) 72.07 ( -0.95%) Min SeqRead-MB/sec-4 75.13 ( 0.00%) 74.92 ( -0.28%) Min SeqRead-MB/sec-8 64.91 ( 0.00%) 65.19 ( 0.43%) Min SeqRead-MB/sec-16 62.24 ( 0.00%) 62.22 ( -0.03%) Min RandRead-MB/sec-1 0.88 ( 0.00%) 0.88 ( 0.00%) Min RandRead-MB/sec-2 0.95 ( 0.00%) 0.92 ( -3.16%) Min RandRead-MB/sec-4 1.43 ( 0.00%) 1.34 ( -6.29%) Min RandRead-MB/sec-8 1.61 ( 0.00%) 1.60 ( -0.62%) Min RandRead-MB/sec-16 1.80 ( 0.00%) 1.90 ( 5.56%) Min SeqWrite-MB/sec-1 76.41 ( 0.00%) 76.85 ( 0.58%) Min SeqWrite-MB/sec-2 74.11 ( 0.00%) 73.54 ( -0.77%) Min SeqWrite-MB/sec-4 80.05 ( 0.00%) 80.13 ( 0.10%) Min SeqWrite-MB/sec-8 72.88 ( 0.00%) 73.20 ( 0.44%) Min SeqWrite-MB/sec-16 75.91 ( 0.00%) 76.44 ( 0.70%) Min RandWrite-MB/sec-1 1.18 ( 0.00%) 1.14 ( -3.39%) Min RandWrite-MB/sec-2 1.02 ( 0.00%) 1.03 ( 0.98%) Min RandWrite-MB/sec-4 1.05 ( 0.00%) 0.98 ( -6.67%) Min RandWrite-MB/sec-8 0.89 ( 0.00%) 0.92 ( 3.37%) Min RandWrite-MB/sec-16 0.92 ( 0.00%) 0.93 ( 1.09%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 approx-v9 User 645.72 525.90 System 403.85 331.75 Elapsed 6795.36 6783.67 This shows that the series has little or not impact on tiobench which is desirable and a reduction in system CPU usage. It indicates that the fair zone allocation policy was removed in a manner that didn't reintroduce one class of page aging bug. There were only minor differences in overall reclaim activity 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Minor Faults 645838 647465 Major Faults 573 640 Swap Ins 0 0 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 46041453 44190646 Normal allocs 78053072 79887245 Movable allocs 0 0 Allocation stalls 24 67 Stall zone DMA 0 0 Stall zone DMA32 0 0 Stall zone Normal 0 2 Stall zone HighMem 0 0 Stall zone Movable 0 65 Direct pages scanned 10969 30609 Kswapd pages scanned 93375144 93492094 Kswapd pages reclaimed 93372243 93489370 Direct pages reclaimed 10969 30609 Kswapd efficiency 99% 99% Kswapd velocity 13741.015 13781.934 Direct efficiency 100% 100% Direct velocity 1.614 4.512 Percentage direct scans 0% 0% kswapd activity was roughly comparable. There were differences in direct reclaim activity but negligible in the context of the overall workload (velocity of 4 pages per second with the patches applied, 1.6 pages per second in the baseline kernel). pgbench read-only large configuration on ext4 --------------------------------------------- pgbench is a database benchmark that can be sensitive to page reclaim decisions. This also checks if removing the fair zone allocation policy is safe pgbench Transactions 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Hmean 1 188.26 ( 0.00%) 189.78 ( 0.81%) Hmean 5 330.66 ( 0.00%) 328.69 ( -0.59%) Hmean 12 370.32 ( 0.00%) 380.72 ( 2.81%) Hmean 21 368.89 ( 0.00%) 369.00 ( 0.03%) Hmean 30 382.14 ( 0.00%) 360.89 ( -5.56%) Hmean 32 428.87 ( 0.00%) 432.96 ( 0.95%) Negligible differences again. As with tiobench, overall reclaim activity was comparable. bonnie++ on ext4 ---------------- No interesting performance difference, negligible differences on reclaim stats. paralleldd on ext4 ------------------ This workload uses varying numbers of dd instances to read large amounts of data from disk. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Amean Elapsd-1 186.04 ( 0.00%) 189.41 ( -1.82%) Amean Elapsd-3 192.27 ( 0.00%) 191.38 ( 0.46%) Amean Elapsd-5 185.21 ( 0.00%) 182.75 ( 1.33%) Amean Elapsd-7 183.71 ( 0.00%) 182.11 ( 0.87%) Amean Elapsd-12 180.96 ( 0.00%) 181.58 ( -0.35%) Amean Elapsd-16 181.36 ( 0.00%) 183.72 ( -1.30%) 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v9 User 1548.01 1552.44 System 8609.71 8515.08 Elapsed 3587.10 3594.54 There is little or no change in performance but some drop in system CPU usage. 4.7.0-rc3 4.7.0-rc3 mmotm-20160623 nodelru-v9 Minor Faults 362662 367360 Major Faults 1204 1143 Swap Ins 22 0 Swap Outs 2855 1029 DMA allocs 0 0 DMA32 allocs 31409797 28837521 Normal allocs 46611853 49231282 Movable allocs 0 0 Direct pages scanned 0 0 Kswapd pages scanned 40845270 40869088 Kswapd pages reclaimed 40830976 40855294 Direct pages reclaimed 0 0 Kswapd efficiency 99% 99% Kswapd velocity 11386.711 11369.769 Direct efficiency 100% 100% Direct velocity 0.000 0.000 Percentage direct scans 0% 0% Page writes by reclaim 2855 1029 Page writes file 0 0 Page writes anon 2855 1029 Page reclaim immediate 771 1628 Sector Reads 293312636 293536360 Sector Writes 18213568 18186480 Page rescued immediate 0 0 Slabs scanned 128257 132747 Direct inode steals 181 56 Kswapd inode steals 59 1131 It basically shows that kswapd was active at roughly the same rate in both kernels. There was also comparable slab scanning activity and direct reclaim was avoided in both cases. There appears to be a large difference in numbers of inodes reclaimed but the workload has few active inodes and is likely a timing artifact. stutter ------- stutter simulates a simple workload. One part uses a lot of anonymous memory, a second measures mmap latency and a third copies a large file. The primary metric is checking for mmap latency. stutter 4.7.0-rc4 4.7.0-rc4 mmotm-20160623 nodelru-v8 Min mmap 16.6283 ( 0.00%) 13.4258 ( 19.26%) 1st-qrtle mmap 54.7570 ( 0.00%) 34.9121 ( 36.24%) 2nd-qrtle mmap 57.3163 ( 0.00%) 46.1147 ( 19.54%) 3rd-qrtle mmap 58.9976 ( 0.00%) 47.1882 ( 20.02%) Max-90% mmap 59.7433 ( 0.00%) 47.4453 ( 20.58%) Max-93% mmap 60.1298 ( 0.00%) 47.6037 ( 20.83%) Max-95% mmap 73.4112 ( 0.00%) 82.8719 (-12.89%) Max-99% mmap 92.8542 ( 0.00%) 88.8870 ( 4.27%) Max mmap 1440.6569 ( 0.00%) 121.4201 ( 91.57%) Mean mmap 59.3493 ( 0.00%) 42.2991 ( 28.73%) Best99%Mean mmap 57.2121 ( 0.00%) 41.8207 ( 26.90%) Best95%Mean mmap 55.9113 ( 0.00%) 39.9620 ( 28.53%) Best90%Mean mmap 55.6199 ( 0.00%) 39.3124 ( 29.32%) Best50%Mean mmap 53.2183 ( 0.00%) 33.1307 ( 37.75%) Best10%Mean mmap 45.9842 ( 0.00%) 20.4040 ( 55.63%) Best5%Mean mmap 43.2256 ( 0.00%) 17.9654 ( 58.44%) Best1%Mean mmap 32.9388 ( 0.00%) 16.6875 ( 49.34%) This shows a number of improvements with the worst-case outlier greatly improved. Some of the vmstats are interesting 4.7.0-rc4 4.7.0-rc4 mmotm-20160623nodelru-v8 Swap Ins 163 502 Swap Outs 0 0 DMA allocs 0 0 DMA32 allocs 618719206 1381662383 Normal allocs 891235743 564138421 Movable allocs 0 0 Allocation stalls 2603 1 Direct pages scanned 216787 2 Kswapd pages scanned 50719775 41778378 Kswapd pages reclaimed 41541765 41777639 Direct pages reclaimed 209159 0 Kswapd efficiency 81% 99% Kswapd velocity 16859.554 14329.059 Direct efficiency 96% 0% Direct velocity 72.061 0.001 Percentage direct scans 0% 0% Page writes by reclaim 6215049 0 Page writes file 6215049 0 Page writes anon 0 0 Page reclaim immediate 70673 90 Sector Reads 81940800 81680456 Sector Writes 100158984 98816036 Page rescued immediate 0 0 Slabs scanned 1366954 22683 While this is not guaranteed in all cases, this particular test showed a large reduction in direct reclaim activity. It's also worth noting that no page writes were issued from reclaim context. This series is not without its hazards. There are at least three areas that I'm concerned with even though I could not reproduce any problems in that area. 1. Reclaim/compaction is going to be affected because the amount of reclaim is no longer targetted at a specific zone. Compaction works on a per-zone basis so there is no guarantee that reclaiming a few THP's worth page pages will have a positive impact on compaction success rates. 2. The Slab/LRU reclaim ratio is affected because the frequency the shrinkers are called is now different. This may or may not be a problem but if it is, it'll be because shrinkers are not called enough and some balancing is required. 3. The anon/file reclaim ratio may be affected. Pages about to be dirtied are distributed between zones and the fair zone allocation policy used to do something very similar for anon. The distribution is now different but not necessarily in any way that matters but it's still worth bearing in mind. VM statistic counters for reclaim decisions are zone-based. If the kernel is to reclaim on a per-node basis then we need to track per-node statistics but there is no infrastructure for that. The most notable change is that the old node_page_state is renamed to sum_zone_node_page_state. The new node_page_state takes a pglist_data and uses per-node stats but none exist yet. There is some renaming such as vm_stat to vm_zone_stat and the addition of vm_node_stat and the renaming of mod_state to mod_zone_state. Otherwise, this is mostly a mechanical patch with no functional change. There is a lot of similarity between the node and zone helpers which is unfortunate but there was no obvious way of reusing the code and maintaining type safety. Link: http://lkml.kernel.org/r/1467970510-21195-2-git-send-email-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Rik van Riel <riel@surriel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Minchan Kim <minchan@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-29 06:45:24 +08:00
sum_zone_node_page_state(nid, i));
for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++)
n += sprintf(buf+n, "%s %lu\n",
vmstat_text[i + NR_VM_ZONE_STAT_ITEMS],
node_page_state(pgdat, i));
return n;
}
static DEVICE_ATTR(vmstat, S_IRUGO, node_read_vmstat, NULL);
static ssize_t node_read_distance(struct device *dev,
struct device_attribute *attr, char *buf)
{
int nid = dev->id;
int len = 0;
int i;
/*
* buf is currently PAGE_SIZE in length and each node needs 4 chars
* at the most (distance + space or newline).
*/
BUILD_BUG_ON(MAX_NUMNODES * 4 > PAGE_SIZE);
for_each_online_node(i)
len += sprintf(buf + len, "%s%d", i ? " " : "", node_distance(nid, i));
len += sprintf(buf + len, "\n");
return len;
}
static DEVICE_ATTR(distance, S_IRUGO, node_read_distance, NULL);
static struct attribute *node_dev_attrs[] = {
&dev_attr_cpumap.attr,
&dev_attr_cpulist.attr,
&dev_attr_meminfo.attr,
&dev_attr_numastat.attr,
&dev_attr_distance.attr,
&dev_attr_vmstat.attr,
NULL
};
ATTRIBUTE_GROUPS(node_dev);
hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-15 09:58:25 +08:00
#ifdef CONFIG_HUGETLBFS
/*
* hugetlbfs per node attributes registration interface:
* When/if hugetlb[fs] subsystem initializes [sometime after this module],
* it will register its per node attributes for all online nodes with
* memory. It will also call register_hugetlbfs_with_node(), below, to
hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-15 09:58:25 +08:00
* register its attribute registration functions with this node driver.
* Once these hooks have been initialized, the node driver will call into
* the hugetlb module to [un]register attributes for hot-plugged nodes.
*/
static node_registration_func_t __hugetlb_register_node;
static node_registration_func_t __hugetlb_unregister_node;
static inline bool hugetlb_register_node(struct node *node)
hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-15 09:58:25 +08:00
{
if (__hugetlb_register_node &&
node_state(node->dev.id, N_MEMORY)) {
hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-15 09:58:25 +08:00
__hugetlb_register_node(node);
return true;
}
return false;
hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-15 09:58:25 +08:00
}
static inline void hugetlb_unregister_node(struct node *node)
{
if (__hugetlb_unregister_node)
__hugetlb_unregister_node(node);
}
void register_hugetlbfs_with_node(node_registration_func_t doregister,
node_registration_func_t unregister)
{
__hugetlb_register_node = doregister;
__hugetlb_unregister_node = unregister;
}
#else
static inline void hugetlb_register_node(struct node *node) {}
static inline void hugetlb_unregister_node(struct node *node) {}
#endif
static void node_device_release(struct device *dev)
{
struct node *node = to_node(dev);
#if defined(CONFIG_MEMORY_HOTPLUG_SPARSE) && defined(CONFIG_HUGETLBFS)
/*
* We schedule the work only when a memory section is
* onlined/offlined on this node. When we come here,
* all the memory on this node has been offlined,
* so we won't enqueue new work to this work.
*
* The work is using node->node_work, so we should
* flush work before freeing the memory.
*/
flush_work(&node->node_work);
#endif
kfree(node);
}
/*
* register_node - Setup a sysfs device for a node.
* @num - Node number to use when creating the device.
*
* Initialize and register the node device.
*/
static int register_node(struct node *node, int num, struct node *parent)
{
int error;
node->dev.id = num;
node->dev.bus = &node_subsys;
node->dev.release = node_device_release;
node->dev.groups = node_dev_groups;
error = device_register(&node->dev);
if (!error){
hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-15 09:58:25 +08:00
hugetlb_register_node(node);
compaction_register_node(node);
}
return error;
}
/**
* unregister_node - unregister a node device
* @node: node going away
*
* Unregisters a node device @node. All the devices on the node must be
* unregistered before calling this function.
*/
void unregister_node(struct node *node)
{
hugetlb_unregister_node(node); /* no-op, if memoryless node */
device_unregister(&node->dev);
}
struct node *node_devices[MAX_NUMNODES];
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
/*
* register cpu under node
*/
int register_cpu_under_node(unsigned int cpu, unsigned int nid)
{
int ret;
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
struct device *obj;
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
if (!node_online(nid))
return 0;
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
obj = get_cpu_device(cpu);
if (!obj)
return 0;
ret = sysfs_create_link(&node_devices[nid]->dev.kobj,
&obj->kobj,
kobject_name(&obj->kobj));
if (ret)
return ret;
return sysfs_create_link(&obj->kobj,
&node_devices[nid]->dev.kobj,
kobject_name(&node_devices[nid]->dev.kobj));
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
}
int unregister_cpu_under_node(unsigned int cpu, unsigned int nid)
{
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
struct device *obj;
if (!node_online(nid))
return 0;
cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem This moves the 'cpu sysdev_class' over to a regular 'cpu' subsystem and converts the devices to regular devices. The sysdev drivers are implemented as subsystem interfaces now. After all sysdev classes are ported to regular driver core entities, the sysdev implementation will be entirely removed from the kernel. Userspace relies on events and generic sysfs subsystem infrastructure from sysdev devices, which are made available with this conversion. Cc: Haavard Skinnemoen <hskinnemoen@gmail.com> Cc: Hans-Christian Egtvedt <egtvedt@samfundet.no> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Borislav Petkov <bp@amd64.org> Cc: Tigran Aivazian <tigran@aivazian.fsnet.co.uk> Cc: Len Brown <lenb@kernel.org> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Dave Jones <davej@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-12-22 06:29:42 +08:00
obj = get_cpu_device(cpu);
if (!obj)
return 0;
sysfs_remove_link(&node_devices[nid]->dev.kobj,
kobject_name(&obj->kobj));
sysfs_remove_link(&obj->kobj,
kobject_name(&node_devices[nid]->dev.kobj));
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
return 0;
}
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
#define page_initialized(page) (page->lru.next)
static int __ref get_nid_for_pfn(unsigned long pfn)
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
{
struct page *page;
if (!pfn_valid_within(pfn))
return -1;
#ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
if (system_state == SYSTEM_BOOTING)
return early_pfn_to_nid(pfn);
#endif
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
page = pfn_to_page(pfn);
if (!page_initialized(page))
return -1;
return pfn_to_nid(pfn);
}
/* register memory section under specified node if it spans that node */
int register_mem_sect_under_node(struct memory_block *mem_blk, int nid)
{
int ret;
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
unsigned long pfn, sect_start_pfn, sect_end_pfn;
if (!mem_blk)
return -EFAULT;
if (!node_online(nid))
return 0;
sect_start_pfn = section_nr_to_pfn(mem_blk->start_section_nr);
sect_end_pfn = section_nr_to_pfn(mem_blk->end_section_nr);
sect_end_pfn += PAGES_PER_SECTION - 1;
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
for (pfn = sect_start_pfn; pfn <= sect_end_pfn; pfn++) {
int page_nid;
mm: check if section present during memory block registering Tony Luck found on his setup, if memory block size 512M will cause crash during booting. BUG: unable to handle kernel paging request at ffffea0074000020 IP: get_nid_for_pfn+0x17/0x40 PGD 128ffcb067 PUD 128ffc9067 PMD 0 Oops: 0000 [#1] SMP Modules linked in: CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.2.0-rc8 #1 ... Call Trace: ? register_mem_sect_under_node+0x66/0xe0 register_one_node+0x17b/0x240 ? pci_iommu_alloc+0x6e/0x6e topology_init+0x3c/0x95 do_one_initcall+0xcd/0x1f0 The system has non continuous RAM address: BIOS-e820: [mem 0x0000001300000000-0x0000001cffffffff] usable BIOS-e820: [mem 0x0000001d70000000-0x0000001ec7ffefff] usable BIOS-e820: [mem 0x0000001f00000000-0x0000002bffffffff] usable BIOS-e820: [mem 0x0000002c18000000-0x0000002d6fffefff] usable BIOS-e820: [mem 0x0000002e00000000-0x00000039ffffffff] usable So there are start sections in memory block not present. For example: memory block : [0x2c18000000, 0x2c20000000) 512M first three sections are not present. The current register_mem_sect_under_node() assume first section is present, but memory block section number range [start_section_nr, end_section_nr] would include not present section. For arch that support vmemmap, we don't setup memmap for struct page area within not present sections area. So skip the pfn range that belong to absent section. [akpm@linux-foundation.org: simplification] [rientjes@google.com: more simplification] Fixes: bdee237c0343 ("x86: mm: Use 2GB memory block size on large memory x86-64 systems") Fixes: 982792c782ef ("x86, mm: probe memory block size for generic x86 64bit") Signed-off-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: David Rientjes <rientjes@google.com> Reported-by: Tony Luck <tony.luck@intel.com> Tested-by: Tony Luck <tony.luck@intel.com> Cc: Greg KH <greg@kroah.com> Cc: Ingo Molnar <mingo@elte.hu> Tested-by: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> [3.15+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-09-05 06:42:39 +08:00
/*
* memory block could have several absent sections from start.
* skip pfn range from absent section
*/
if (!pfn_present(pfn)) {
pfn = round_down(pfn + PAGES_PER_SECTION,
PAGES_PER_SECTION) - 1;
continue;
}
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
page_nid = get_nid_for_pfn(pfn);
if (page_nid < 0)
continue;
if (page_nid != nid)
continue;
ret = sysfs_create_link_nowarn(&node_devices[nid]->dev.kobj,
&mem_blk->dev.kobj,
kobject_name(&mem_blk->dev.kobj));
if (ret)
return ret;
return sysfs_create_link_nowarn(&mem_blk->dev.kobj,
&node_devices[nid]->dev.kobj,
kobject_name(&node_devices[nid]->dev.kobj));
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
}
/* mem section does not span the specified node */
return 0;
}
/* unregister memory section under all nodes that it spans */
int unregister_mem_sect_under_nodes(struct memory_block *mem_blk,
unsigned long phys_index)
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
{
NODEMASK_ALLOC(nodemask_t, unlinked_nodes, GFP_KERNEL);
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
unsigned long pfn, sect_start_pfn, sect_end_pfn;
if (!mem_blk) {
NODEMASK_FREE(unlinked_nodes);
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
return -EFAULT;
}
if (!unlinked_nodes)
return -ENOMEM;
nodes_clear(*unlinked_nodes);
sect_start_pfn = section_nr_to_pfn(phys_index);
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
sect_end_pfn = sect_start_pfn + PAGES_PER_SECTION - 1;
for (pfn = sect_start_pfn; pfn <= sect_end_pfn; pfn++) {
mm: get_nid_for_pfn() returns int get_nid_for_pfn() returns int Presumably the (nid < 0) case has never happened. We do know that it is happening on one system while creating a symlink for a memory section so it should also happen on the same system if unregister_mem_sect_under_nodes() were called to remove the same symlink. The test was actually added in response to a problem with an earlier version reported by Yasunori Goto where one or more of the leading pages of a memory section on the 2nd node of one of his systems was uninitialized because I believe they coincided with a memory hole. That earlier version did not ignore uninitialized pages and determined the nid by considering only the 1st page of each memory section. This caused the symlink to the 1st memory section on the 2nd node to be incorrectly created in /sys/devices/system/node/node0 instead of /sys/devices/system/node/node1. The problem was fixed by adding the test to skip over uninitialized pages. I suspect we have not seen any reports of the non-removal of a symlink due to the incorrect declaration of the nid variable in unregister_mem_sect_under_nodes() because - systems where a memory section could have an uninitialized range of leading pages are probably rare. - memory remove is probably not done very frequently on the systems that are capable of demonstrating the problem. - lingering symlink(s) that should have been removed may have simply gone unnoticed. [garyhade@us.ibm.com: wrote changelog] Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Cc: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-11 03:55:45 +08:00
int nid;
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
nid = get_nid_for_pfn(pfn);
if (nid < 0)
continue;
if (!node_online(nid))
continue;
if (node_test_and_set(nid, *unlinked_nodes))
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
continue;
sysfs_remove_link(&node_devices[nid]->dev.kobj,
kobject_name(&mem_blk->dev.kobj));
sysfs_remove_link(&mem_blk->dev.kobj,
kobject_name(&node_devices[nid]->dev.kobj));
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
}
NODEMASK_FREE(unlinked_nodes);
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
return 0;
}
static int link_mem_sections(int nid)
{
unsigned long start_pfn = NODE_DATA(nid)->node_start_pfn;
unsigned long end_pfn = start_pfn + NODE_DATA(nid)->node_spanned_pages;
unsigned long pfn;
struct memory_block *mem_blk = NULL;
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
int err = 0;
for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
unsigned long section_nr = pfn_to_section_nr(pfn);
struct mem_section *mem_sect;
int ret;
if (!present_section_nr(section_nr))
continue;
mem_sect = __nr_to_section(section_nr);
drivers/base/memory.c: fix memory_dev_init() long delay One system with 2048g ram, reported soft lockup on recent kernel. [ 34.426749] cpu_dev_init done [ 61.166399] BUG: soft lockup - CPU#0 stuck for 22s! [swapper/0:1] [ 61.166733] Modules linked in: [ 61.166904] irq event stamp: 1935610 [ 61.178431] hardirqs last enabled at (1935609): [<ffffffff81ce8c05>] mutex_lock_nested+0x299/0x2b4 [ 61.178923] hardirqs last disabled at (1935610): [<ffffffff81cf2bab>] apic_timer_interrupt+0x6b/0x80 [ 61.198767] softirqs last enabled at (1935476): [<ffffffff8106e59c>] __do_softirq+0x195/0x1ab [ 61.218604] softirqs last disabled at (1935471): [<ffffffff81cf359c>] call_softirq+0x1c/0x30 [ 61.238408] CPU 0 [ 61.238549] Modules linked in: [ 61.238744] [ 61.238825] Pid: 1, comm: swapper/0 Not tainted 3.3.0-rc1-tip-yh-02076-g962f689-dirty #171 [ 61.278212] RIP: 0010:[<ffffffff810b3e3a>] [<ffffffff810b3e3a>] lock_release+0x90/0x9c [ 61.278627] RSP: 0018:ffff883f64dbfd70 EFLAGS: 00000246 [ 61.298287] RAX: ffff883f64dc0000 RBX: 0000000000000000 RCX: 000000000000008b [ 61.298690] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 [ 61.318383] RBP: ffff883f64dbfda0 R08: 0000000000000001 R09: 000000000000008b [ 61.338215] R10: 0000000000000000 R11: 0000000000000000 R12: ffff883f64dbfd10 [ 61.338610] R13: ffff883f64dc0708 R14: ffff883f64dc0708 R15: ffffffff81095657 [ 61.358299] FS: 0000000000000000(0000) GS:ffff883f7d600000(0000) knlGS:0000000000000000 [ 61.378118] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [ 61.378450] CR2: 0000000000000000 CR3: 00000000024af000 CR4: 00000000000007f0 [ 61.398144] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 61.417918] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 61.418260] Process swapper/0 (pid: 1, threadinfo ffff883f64dbe000, task ffff883f64dc0000) [ 61.445358] Stack: [ 61.445511] 0000000000000002 ffff897f649ba168 ffff883f64dbfe10 ffff88ff64bb57a8 [ 61.458040] 0000000000000000 0000000000000000 ffff883f64dbfdc0 ffffffff81ceb1b4 [ 61.458491] 000000000011608c ffff88ff64bb58a8 ffff883f64dbfdf0 ffffffff81c57638 [ 61.478215] Call Trace: [ 61.478367] [<ffffffff81ceb1b4>] _raw_spin_unlock+0x21/0x2e [ 61.497994] [<ffffffff81c57638>] klist_next+0x9e/0xbc [ 61.498264] [<ffffffff8148ba99>] next_device+0xe/0x1e [ 61.517867] [<ffffffff8148c0cc>] subsys_find_device_by_id+0xb7/0xd6 [ 61.518197] [<ffffffff81498846>] find_memory_block_hinted+0x3d/0x66 [ 61.537927] [<ffffffff8149887f>] find_memory_block+0x10/0x12 [ 61.538193] [<ffffffff814988b6>] add_memory_section+0x35/0x9e [ 61.557932] [<ffffffff827fecef>] memory_dev_init+0x68/0xda [ 61.558227] [<ffffffff827fec01>] driver_init+0x97/0xa7 [ 61.577853] [<ffffffff827cdf3c>] kernel_init+0xf6/0x1c0 [ 61.578140] [<ffffffff81cf34a4>] kernel_thread_helper+0x4/0x10 [ 61.597850] [<ffffffff81ceb59d>] ? retint_restore_args+0xe/0xe [ 61.598144] [<ffffffff827cde46>] ? start_kernel+0x3ab/0x3ab [ 61.617826] [<ffffffff81cf34a0>] ? gs_change+0xb/0xb [ 61.618060] Code: 10 48 83 3b 00 eb e8 4c 89 f2 44 89 fe 4c 89 ef e8 e1 fe ff ff 65 48 8b 04 25 40 bc 00 00 c7 80 cc 06 00 00 00 00 00 00 41 54 9d <5e> 5b 41 5c 41 5d 41 5e 41 5f 5d c3 55 48 89 e5 41 57 41 89 cf [ 89.285380] memory_dev_init done Finally it takes about 55s to create 16400 memory entries. Root cause: for x86_64, 2048g (with 2g hole at [2g,4g), and TOP2 will be 2050g), will have 16400 memory block. find_memory_block/subsys_find_device_by_id will be expensive with that many entries. Actually, we don't need to find that memory block for BOOT path. Skip that finding make it get back to normal. [ 34.466696] cpu_dev_init done [ 35.290080] memory_dev_init done Also solved the delay with topology_init when sections_per_block is not 1. Signed-off-by: Yinghai Lu <yinghai@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Nathan Fontenot <nfont@austin.ibm.com> Cc: Robin Holt <holt@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-01-31 05:57:12 +08:00
/* same memblock ? */
if (mem_blk)
if ((section_nr >= mem_blk->start_section_nr) &&
(section_nr <= mem_blk->end_section_nr))
continue;
mem_blk = find_memory_block_hinted(mem_sect, mem_blk);
drivers/base/memory.c: fix memory_dev_init() long delay One system with 2048g ram, reported soft lockup on recent kernel. [ 34.426749] cpu_dev_init done [ 61.166399] BUG: soft lockup - CPU#0 stuck for 22s! [swapper/0:1] [ 61.166733] Modules linked in: [ 61.166904] irq event stamp: 1935610 [ 61.178431] hardirqs last enabled at (1935609): [<ffffffff81ce8c05>] mutex_lock_nested+0x299/0x2b4 [ 61.178923] hardirqs last disabled at (1935610): [<ffffffff81cf2bab>] apic_timer_interrupt+0x6b/0x80 [ 61.198767] softirqs last enabled at (1935476): [<ffffffff8106e59c>] __do_softirq+0x195/0x1ab [ 61.218604] softirqs last disabled at (1935471): [<ffffffff81cf359c>] call_softirq+0x1c/0x30 [ 61.238408] CPU 0 [ 61.238549] Modules linked in: [ 61.238744] [ 61.238825] Pid: 1, comm: swapper/0 Not tainted 3.3.0-rc1-tip-yh-02076-g962f689-dirty #171 [ 61.278212] RIP: 0010:[<ffffffff810b3e3a>] [<ffffffff810b3e3a>] lock_release+0x90/0x9c [ 61.278627] RSP: 0018:ffff883f64dbfd70 EFLAGS: 00000246 [ 61.298287] RAX: ffff883f64dc0000 RBX: 0000000000000000 RCX: 000000000000008b [ 61.298690] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 [ 61.318383] RBP: ffff883f64dbfda0 R08: 0000000000000001 R09: 000000000000008b [ 61.338215] R10: 0000000000000000 R11: 0000000000000000 R12: ffff883f64dbfd10 [ 61.338610] R13: ffff883f64dc0708 R14: ffff883f64dc0708 R15: ffffffff81095657 [ 61.358299] FS: 0000000000000000(0000) GS:ffff883f7d600000(0000) knlGS:0000000000000000 [ 61.378118] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [ 61.378450] CR2: 0000000000000000 CR3: 00000000024af000 CR4: 00000000000007f0 [ 61.398144] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 61.417918] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 61.418260] Process swapper/0 (pid: 1, threadinfo ffff883f64dbe000, task ffff883f64dc0000) [ 61.445358] Stack: [ 61.445511] 0000000000000002 ffff897f649ba168 ffff883f64dbfe10 ffff88ff64bb57a8 [ 61.458040] 0000000000000000 0000000000000000 ffff883f64dbfdc0 ffffffff81ceb1b4 [ 61.458491] 000000000011608c ffff88ff64bb58a8 ffff883f64dbfdf0 ffffffff81c57638 [ 61.478215] Call Trace: [ 61.478367] [<ffffffff81ceb1b4>] _raw_spin_unlock+0x21/0x2e [ 61.497994] [<ffffffff81c57638>] klist_next+0x9e/0xbc [ 61.498264] [<ffffffff8148ba99>] next_device+0xe/0x1e [ 61.517867] [<ffffffff8148c0cc>] subsys_find_device_by_id+0xb7/0xd6 [ 61.518197] [<ffffffff81498846>] find_memory_block_hinted+0x3d/0x66 [ 61.537927] [<ffffffff8149887f>] find_memory_block+0x10/0x12 [ 61.538193] [<ffffffff814988b6>] add_memory_section+0x35/0x9e [ 61.557932] [<ffffffff827fecef>] memory_dev_init+0x68/0xda [ 61.558227] [<ffffffff827fec01>] driver_init+0x97/0xa7 [ 61.577853] [<ffffffff827cdf3c>] kernel_init+0xf6/0x1c0 [ 61.578140] [<ffffffff81cf34a4>] kernel_thread_helper+0x4/0x10 [ 61.597850] [<ffffffff81ceb59d>] ? retint_restore_args+0xe/0xe [ 61.598144] [<ffffffff827cde46>] ? start_kernel+0x3ab/0x3ab [ 61.617826] [<ffffffff81cf34a0>] ? gs_change+0xb/0xb [ 61.618060] Code: 10 48 83 3b 00 eb e8 4c 89 f2 44 89 fe 4c 89 ef e8 e1 fe ff ff 65 48 8b 04 25 40 bc 00 00 c7 80 cc 06 00 00 00 00 00 00 41 54 9d <5e> 5b 41 5c 41 5d 41 5e 41 5f 5d c3 55 48 89 e5 41 57 41 89 cf [ 89.285380] memory_dev_init done Finally it takes about 55s to create 16400 memory entries. Root cause: for x86_64, 2048g (with 2g hole at [2g,4g), and TOP2 will be 2050g), will have 16400 memory block. find_memory_block/subsys_find_device_by_id will be expensive with that many entries. Actually, we don't need to find that memory block for BOOT path. Skip that finding make it get back to normal. [ 34.466696] cpu_dev_init done [ 35.290080] memory_dev_init done Also solved the delay with topology_init when sections_per_block is not 1. Signed-off-by: Yinghai Lu <yinghai@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Nathan Fontenot <nfont@austin.ibm.com> Cc: Robin Holt <holt@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-01-31 05:57:12 +08:00
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
ret = register_mem_sect_under_node(mem_blk, nid);
if (!err)
err = ret;
/* discard ref obtained in find_memory_block() */
}
if (mem_blk)
kobject_put(&mem_blk->dev.kobj);
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
return err;
}
#ifdef CONFIG_HUGETLBFS
/*
* Handle per node hstate attribute [un]registration on transistions
* to/from memoryless state.
*/
static void node_hugetlb_work(struct work_struct *work)
{
struct node *node = container_of(work, struct node, node_work);
/*
* We only get here when a node transitions to/from memoryless state.
* We can detect which transition occurred by examining whether the
* node has memory now. hugetlb_register_node() already check this
* so we try to register the attributes. If that fails, then the
* node has transitioned to memoryless, try to unregister the
* attributes.
*/
if (!hugetlb_register_node(node))
hugetlb_unregister_node(node);
}
static void init_node_hugetlb_work(int nid)
{
INIT_WORK(&node_devices[nid]->node_work, node_hugetlb_work);
}
static int node_memory_callback(struct notifier_block *self,
unsigned long action, void *arg)
{
struct memory_notify *mnb = arg;
int nid = mnb->status_change_nid;
switch (action) {
case MEM_ONLINE:
case MEM_OFFLINE:
/*
* offload per node hstate [un]registration to a work thread
* when transitioning to/from memoryless state.
*/
if (nid != NUMA_NO_NODE)
schedule_work(&node_devices[nid]->node_work);
break;
case MEM_GOING_ONLINE:
case MEM_GOING_OFFLINE:
case MEM_CANCEL_ONLINE:
case MEM_CANCEL_OFFLINE:
default:
break;
}
return NOTIFY_OK;
}
#endif /* CONFIG_HUGETLBFS */
#else /* !CONFIG_MEMORY_HOTPLUG_SPARSE */
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
static int link_mem_sections(int nid) { return 0; }
#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */
#if !defined(CONFIG_MEMORY_HOTPLUG_SPARSE) || \
!defined(CONFIG_HUGETLBFS)
static inline int node_memory_callback(struct notifier_block *self,
unsigned long action, void *arg)
{
return NOTIFY_OK;
}
static void init_node_hugetlb_work(int nid) { }
#endif
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
int register_one_node(int nid)
{
int error = 0;
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
int cpu;
if (node_online(nid)) {
int p_node = parent_node(nid);
struct node *parent = NULL;
if (p_node != nid)
parent = node_devices[p_node];
node_devices[nid] = kzalloc(sizeof(struct node), GFP_KERNEL);
if (!node_devices[nid])
return -ENOMEM;
error = register_node(node_devices[nid], nid, parent);
[PATCH] node hotplug: register cpu: remove node struct With Goto-san's patch, we can add new pgdat/node at runtime. I'm now considering node-hot-add with cpu + memory on ACPI. I found acpi container, which describes node, could evaluate cpu before memory. This means cpu-hot-add occurs before memory hot add. In most part, cpu-hot-add doesn't depend on node hot add. But register_cpu(), which creates symbolic link from node to cpu, requires that node should be onlined before register_cpu(). When a node is onlined, its pgdat should be there. This patch-set holds off creating symbolic link from node to cpu until node is onlined. This removes node arguments from register_cpu(). Now, register_cpu() requires 'struct node' as its argument. But the array of struct node is now unified in driver/base/node.c now (By Goto's node hotplug patch). We can get struct node in generic way. So, this argument is not necessary now. This patch also guarantees add cpu under node only when node is onlined. It is necessary for node-hot-add vs. cpu-hot-add patch following this. Moreover, register_cpu calculates cpu->node_id by cpu_to_node() without regard to its 'struct node *root' argument. This patch removes it. Also modify callers of register_cpu()/unregister_cpu, whose args are changed by register-cpu-remove-node-struct patch. [Brice.Goglin@ens-lyon.org: fix it] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Brice Goglin <Brice.Goglin@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-27 17:53:41 +08:00
/* link cpu under this node */
for_each_present_cpu(cpu) {
if (cpu_to_node(cpu) == nid)
register_cpu_under_node(cpu, nid);
}
mm: show node to memory section relationship with symlinks in sysfs Show node to memory section relationship with symlinks in sysfs Add /sys/devices/system/node/nodeX/memoryY symlinks for all the memory sections located on nodeX. For example: /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 indicates that memory section 135 resides on node1. Also revises documentation to cover this change as well as updating Documentation/ABI/testing/sysfs-devices-memory to include descriptions of memory hotremove files 'phys_device', 'phys_index', and 'state' that were previously not described there. In addition to it always being a good policy to provide users with the maximum possible amount of physical location information for resources that can be hot-added and/or hot-removed, the following are some (but likely not all) of the user benefits provided by this change. Immediate: - Provides information needed to determine the specific node on which a defective DIMM is located. This will reduce system downtime when the node or defective DIMM is swapped out. - Prevents unintended onlining of a memory section that was previously offlined due to a defective DIMM. This could happen during node hot-add when the user or node hot-add assist script onlines _all_ offlined sections due to user or script inability to identify the specific memory sections located on the hot-added node. The consequences of reintroducing the defective memory could be ugly. - Provides information needed to vary the amount and distribution of memory on specific nodes for testing or debugging purposes. Future: - Will provide information needed to identify the memory sections that need to be offlined prior to physical removal of a specific node. Symlink creation during boot was tested on 2-node x86_64, 2-node ppc64, and 2-node ia64 systems. Symlink creation during physical memory hot-add tested on a 2-node x86_64 system. Signed-off-by: Gary Hade <garyhade@us.ibm.com> Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-07 06:39:14 +08:00
/* link memory sections under this node */
error = link_mem_sections(nid);
/* initialize work queue for memory hot plug */
init_node_hugetlb_work(nid);
}
return error;
}
void unregister_one_node(int nid)
{
if (!node_devices[nid])
return;
unregister_node(node_devices[nid]);
node_devices[nid] = NULL;
}
/*
* node states attributes
*/
static ssize_t print_nodes_state(enum node_states state, char *buf)
{
int n;
n = scnprintf(buf, PAGE_SIZE - 1, "%*pbl",
nodemask_pr_args(&node_states[state]));
buf[n++] = '\n';
buf[n] = '\0';
return n;
}
struct node_attr {
struct device_attribute attr;
enum node_states state;
};
static ssize_t show_node_state(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct node_attr *na = container_of(attr, struct node_attr, attr);
return print_nodes_state(na->state, buf);
}
#define _NODE_ATTR(name, state) \
{ __ATTR(name, 0444, show_node_state, NULL), state }
static struct node_attr node_state_attr[] = {
[N_POSSIBLE] = _NODE_ATTR(possible, N_POSSIBLE),
[N_ONLINE] = _NODE_ATTR(online, N_ONLINE),
[N_NORMAL_MEMORY] = _NODE_ATTR(has_normal_memory, N_NORMAL_MEMORY),
#ifdef CONFIG_HIGHMEM
[N_HIGH_MEMORY] = _NODE_ATTR(has_high_memory, N_HIGH_MEMORY),
#endif
#ifdef CONFIG_MOVABLE_NODE
[N_MEMORY] = _NODE_ATTR(has_memory, N_MEMORY),
#endif
[N_CPU] = _NODE_ATTR(has_cpu, N_CPU),
};
static struct attribute *node_state_attrs[] = {
&node_state_attr[N_POSSIBLE].attr.attr,
&node_state_attr[N_ONLINE].attr.attr,
&node_state_attr[N_NORMAL_MEMORY].attr.attr,
#ifdef CONFIG_HIGHMEM
&node_state_attr[N_HIGH_MEMORY].attr.attr,
#endif
#ifdef CONFIG_MOVABLE_NODE
&node_state_attr[N_MEMORY].attr.attr,
#endif
&node_state_attr[N_CPU].attr.attr,
NULL
};
static struct attribute_group memory_root_attr_group = {
.attrs = node_state_attrs,
};
static const struct attribute_group *cpu_root_attr_groups[] = {
&memory_root_attr_group,
NULL,
};
#define NODE_CALLBACK_PRI 2 /* lower than SLAB */
static int __init register_node_type(void)
{
int ret;
BUILD_BUG_ON(ARRAY_SIZE(node_state_attr) != NR_NODE_STATES);
BUILD_BUG_ON(ARRAY_SIZE(node_state_attrs)-1 != NR_NODE_STATES);
ret = subsys_system_register(&node_subsys, cpu_root_attr_groups);
if (!ret) {
static struct notifier_block node_memory_callback_nb = {
.notifier_call = node_memory_callback,
.priority = NODE_CALLBACK_PRI,
};
register_hotmemory_notifier(&node_memory_callback_nb);
}
/*
* Note: we're not going to unregister the node class if we fail
* to register the node state class attribute files.
*/
return ret;
}
postcore_initcall(register_node_type);